2022. December 5.

Do you care about about climate change?

Have you been following COP27?

Do you think international climate action will deliver results?

Background: Climate change is a global issue where international solutions are needed to deliver results

‘Tragedy of the commons’: limited incentives for any single actor to stop overusing1…

  • Global common resources, such as air, water, fisheries have been overused and polluted for the past decades

  • Mismatched costs and benefits: resource users do not bear the costs of pollution or overuse directly, or immediately

  • Traditional solutions, especially local ones, cannot effectively limit global warning as cross-country resources are affected

  • International cooperation on climate change is necessary to limit its effects

…but international solutions so far have been lacking in providing effective solutions

  • Multiple times, global or regional solutions have been attempted to limit pollution and curtail global warning

  • But with limited enforcement power, these solutions so far have not achieved a significant reduction in greenhouse gas emissions2

  • The latest climate agreements, known as the Paris Agreement and the newly decided ‘Damage and Loss’ initiative at COP27 try to provide a framework to reduce emissions and mitigate existing consequences3

Background: Literature covers opinions about climate action, but the relationship with other social-political views is less explored

Figure 1: Illustrative example of literature

People’s views on climate action have been investigated before:

  • Studies typically show people are concerned about climate change (e.g., ~60%+ of Americans are somewhat concerned4), that younger generations are more concerned5, or that income, education or political views matter6

  • Yet how different perspectives potentially influence each other is less commonly explored

  • People’s views and opinions on climate action or the results of those actions might be influenced by, or correlated with other views they hold about the world - for example, how they think about diversity, their own country, or democracy in general

Research: Our research focuses on how people’s social-political views interact with their views on international climate action

Our research question is two-fold:

Question 1: Is people’s satisfaction with democracy in their own country or their views on diversity influences how they perceive international climate action?

  • Y: People’s belief if international climate actions can solve climate change

  • X1: People’s satisfaction with democracy in their own country, understood as a binary indicator (satisfied / not satisfied)

  • X2: People’s opinion on whether diversity makes the world a better place (i.e., having people with different backgrounds, such as different ethnic groups, religions and races living in their country is good or bad)

  • Control variables: age, gender, income, education, climate concern, race

Question 2: Does this relationship differ across countries in the developed world?

Hypothesis: We expect a significant positive relationship for both independent variables across countires

Research: We narrowed down our scope of analysis to six countries across four continents based on three selective criteria

1. Countries with outsized economic impact / outsized polluters

Measurement: Top10 global polluters7

  • United States

  • Germany

  • South Korea

2. Countries that are considered to be at the forefront of climate change action

Measurement: per capita fossil fuel usage has been decreasing 8

  • Australia

  • Sweden

  • The Netherlands

3. Geographical diversity within the above two criteria

Data: We selected two PEW datasets that cover our research topic and all countries, and were conducted at a similar time

1. Global Attitudes Survey, Spring 2021, Pew Research Center

  • Survey method: phone
  • Participants: 16,254 adults in 16 advanced economies (Canada, Belgium, France, Germany, Greece, Italy, the Netherlands, Spain, Sweden, UK, Australia, Japan, New Zealand, Singapore, S-Korea, Taiwan)
  • Sampling: Random Digit Dialing
    • Availability of landline and mobile infrastructure, telephone polls as an established feature of the local opinion research to ensure nationally representative, probability-based designs
  • Data quality checks: access to technology and cultural considerations

2. American Trends Panel, February 2021, Pew Research Center

  • Survey method: online, in English & Spanish
  • Participants: 2,596 US adults (tablet & wireless internet provided if needed)
  • Sampling: stratified random sampling to ensure that nearly all adults have a chance of selection
  • Data quality checks: removal of respondents with clear patterns of satisficing
  • Incentives: a post-paid check or Amazon gift codes ($5-20); differential amount according to survey response propensities

Method: Achieving consistency across countries and the two datasets proved a significant challenge during data cleaning I.

I. Data cleaning

 1. Subset only to countries that relate to the scope of our analysis

 2. To allow for comparability across countries:

  • Cleaning data within the focal country of analysis

    • select the single best, appropriate measure, where there are multiple variables of the same concept (e.g. Political ideology: continuum vs. specific party)

  • Cleaning data across countries

    • standardize the survey response items

  • Cleaning within the Global dataset (but across AU, DE, NL, KR, SE)

    • Education example: required greater understanding as credentials/qualifications were adapted to the national context / were categorized differently
      • Additional cleaning was required for the AUS sample, where missing data consisted of both true NAs and primary/secondary school graduates
    • Income example: different currencies / numbers of income categories that also differed in range / timeframes (annual vs. monthly income)

Method: Education recoding example I.

Raw code

Figure 2: Illustrative example of three countries, raw values

Initial recoding

Figure 3: Illustrative example of three countries, initial recoding

Method: Education recoding example II.

Final recoded version

Figure 4: Illustrative example of three countries, final recoded version

Method: Income recoding example I.

Raw code

Figure 5: Illustrative example of four countries, raw values

Method: Income recoding example II.

Final recoded version

Figure 5: Illustrative example of four countries, final recoded version

Method: Achieving consistency across countries and the two datasets proved a significant challenge during data cleaning II.

  • Cleaning between the Global & US dataset
    • e.g. Political Ideology: 7-point vs. 5-point scale; Left-Right vs. Right-Left spectrum
    • e.g. Age: actual age (straight method) vs. age category
    • e.g. Attitudes toward diversity: binary vs. 3 types of response categories
    • e.g. Sex: different numeric values assigned to females and males

Thus, recode:

  • EDUCATION - BA or higher = 1; No BA = 0
  • INCOME - equal to or above the national median = 1; below the median = 0
  • DIVERSITY_GOOD - Diversity makes a better place to live = 1; Otherwise = 0
  • POLITICAL_ID - left = 1; moderate = 2; right = 3
  • AGE - 18-29 = 1; 30-49 = 2; 50-64 = 3; 65+ = 4
  • FEMALE - female = 1; non-female = 0

Method: After data cleaning and merging, logistic regression was run on the six countries

 3. To perform a logistic regression, create a binary variable for the Y variable (CLIMATE_CONFIDENCE)

  • CLIMATE_CONFIDENCE - Confident that international climate action will reduce the effects of climate change = 1; Not confident = 0

 4. Convert ‘Don’t Know’ and ‘Refused’ to NAs

 5. Change the order of the columns so that we can identify the respondents’ nationality at first glance. Then, select only the variables relevant to our analysis.

 6. Merge the Global dataset and the US dataset

II. Regression modeling (for the 6 countries)

  • Defining our hypothesized multiple logistic regression models

  • Consolidating and interpreting the coefficient estimates of the regression tables and their statistical significance

III. Converting the log odds into predicted probabilities to allow for intuitive understanding of the relationship

Results: Both diversity and satifaction with democracy had a significant relationship with confidence in int. climate action

Figure 6: regression results across the six country models



Who tends to be more optimistic about the effectiveness of international climate action?

  • Overall, those with a) positive attitudes toward social diversity and b) low satisfaction with the functioning of democracy in one’s own country believe in the power of multilateral solutions
  • The latter contradicts our initial hypothesis of a positive relationship between satisfaction with democracy and confidence in solving climate change through collaborative efforts

Potential explanations for the contradictory finding

  • The more skeptical one is of the ability of domestic institutions and political systems to handle societal problems internally, the more one believes that international undertaking is a viable solution to tackling these problems

Results: In Australia, the relationship between diversity and confidence in international climate action was not supported

Results: Differences regarding predicted probabilities on democracy were the greatest in US and NL

Figure 8: Predicted probabilities, democracy perceptions





Differences in the probability that an individual with low vs. high satisfaction with democracy would be confident about int. climate action:

(provided that control variables are set to their median levels)

  • The USA (48% points) and the Netherlands (46% points) demonstrated the largest difference among the two groups of individuals
  • Compared to Australia (21% points), Sweden (26% points), Germany (28% points), Korea (12% points)

Note: Small differences in the predicted probabilities for the two groups in Korea is consistent with our findings that the effect of satisfaction with democracy was significant (at 0.1 significance level, as opposed to the conventional 0.05 level) using the p-value approach, and that we failed to reject the null hypothesis using the CI approach - by an extremely narrow margin

Results: Predicted probabilities on diversity: The Swedes were less optimistic about tackling climate change through multilateralism

Figure 8: Predicted probabilities, diversity







Possible explanations for the Swedes’ skepticism

 1. The difficulty of getting the world’s largest powers involved in combating climate change and set an example for the others, despite the importance of collective effort.

  • Although the Swedes tended to believe that their country (73%) and the UN (58%) was doing a good job of handling climate change, they tended to be the most critical of US’ (75%) and China’s (89%) handling of the issue, out of all countries surveyed.
  • e.g. US’ decision to exit and rejoin the Paris Agreement

 2. The belief that international efforts to take climate action such as the COP26 are more of an “empty talk” and “empty promises”

Results: Predicted probabilities on diversity: Koreans tended to be more optimistic regarding diversity

Figure 8: Predicted probabilities, diversity







Possible explanations for optimism among Koreans

Korea’s increased presence in the multilateral stage and the pledge of “action and solidarity” at the COP26, APEC, and G20 summit, where the climate crisis was high on the agenda

  • Delivered a key note speech at the UN climate summit COP26 (2021) and the APEC CEO summit (2021)
  • At COP26, announced its decision to join the Global Methane Pledge, a global pact to cut methane emissions by 30 percent by 2030 (in which the EU, UK, and US also participate); plans to join global efforts to reduce the use of coal; commitment to cooperation in forest restoration
  • Korea was to be the host of the World Forestry Congress (2022)

Conclusions: The cross-country analysis provided interesting insight into societies that can be the basis of further studies

Overall, those with a) positive attitudes toward social diversity and b) low satisfaction with the functioning of democracy in one’s own country believe in the power of multilateral solutions

  • This holds true across all investigated countries except Australia, where diversity was not significant
  • Possibly, the more skeptical one is of the ability of domestic institutions and political systems to handle societal problems internally, the more one believes that international undertaking is a viable solution to tackling these problems

Cross-country comparison revealed differences between countries and how individuals feel about democracy and diversity in relation to international climate action

  • Regarding democracy, the variance was greater than diversity
  • Differences might be due to each country’s specific institutions, their ability to influence global decision-making, or how individuals view other international actors such as the EU, the USA or China

Conclusions: Our study was limited in its scope to developed countries and to a few socio-political indicators

Limitations on geography

  • We only had data on developed countries - including developing countries would make the study more meaningful
  • We limited the scope of analysis to six countries, but the rest of the countries might be also analyzed

Limitations on independent variables and controls

  • More independent variables might have been included to investigate other, potentially relevant socio-political views
  • Further control variables might have been identified with a more extensive literature review

Limitations on time

  • Different time periods could also be investigated to see if e.g., we see a different reaction right after major global events, such as after the Paris Agreement or after the COP27
  • Different climate-related events could be compared to see if there are differences
  • Non-climate related events might also be taken into account such as respective elections

Next steps: Further countries and variables could be included in the analysis to provide a more holistic picture

Further analysis

  • The results can be further analyzed to understand outliers and reasons behind differences regarding each country
  • Control variable differences could be further explored, and the models potentially adjusted to include / exclude some specific variables



Comparative review

  • Results could be compared with other, similar studies
  • Similar analysis could be conducted on different datasets covering the same countries

Appendix

Appendix: We included age, income, gender, education and climate concern as our control variables, and race for the United States

Age

  • Using data from Climate Change in the American Mind, Yale researchers found that younger Republicans are more climate aware; PEW researched Gen Z activism in relation to climate change vs. older generation and found age relevant; Dijkstra, E.M. & Goedhart, M.J. found that older students score lower on climate attitudes; Poortinga et al. found that climate scepticism is stronger in older people

Income

  • A PEW report found that higher income individuals living in developed countries are willing to make some changes to combat climate change; Poortinga et al. found that climate scepticism is more prevalent amongst individuals from lower socio-economic background

Political affiliation

  • A report by PEW research found that party affiliation in America defines attitudes about climate change, and another report found similar results internationally; Poortinga et al. found that is more prevalent amongst conservatives

Education

  • Rode et al. found in a meta-analysis that education influences climate change attitudes

Gender

  • Rode et al. found in a meta-analysis that gender influences climate change attitudes

Appendix: Results of regression

The effect of diversity and democracy perceptions on confidence in international climate action
Dependent variable:
CLIMATE_CONFIDENCE
AUSTRALIA KOREA GERMANY NETHERLANDS SWEDEN USA
(1) (2) (3) (4) (5) (6)
DIVERSITY_GOOD 0.344 0.416** 0.463** 0.424*** 0.560*** 0.568***
(0.246) (0.163) (0.186) (0.162) (0.198) (0.188)
SATISFIED_DEMOCRACY -0.288*** -0.184* -0.393*** -0.655*** -0.438*** -0.720***
(0.090) (0.097) (0.079) (0.087) (0.098) (0.064)
CLIMATE_CONCERN -0.410*** -0.472*** -0.082 -0.172** -0.128 -0.795***
(0.083) (0.107) (0.079) (0.084) (0.081) (0.059)
FEMALE 0.308** 0.305* 0.181 0.097 -0.141 0.288***
(0.151) (0.182) (0.150) (0.151) (0.144) (0.099)
AGECAT -0.021 0.113 0.230*** 0.123* -0.086 -0.056
(0.076) (0.091) (0.074) (0.071) (0.072) (0.050)
EDUCATION -0.355** 0.156 -0.184 -0.625*** -0.134 -0.095
(0.162) (0.169) (0.163) (0.163) (0.157) (0.108)
POLITICAL_ID -0.074 -0.083 0.128 0.170* 0.081 -0.346***
(0.110) (0.117) (0.104) (0.098) (0.095) (0.073)
INCOME 0.059 0.055 -0.406*** -0.050 -0.481*** -0.047
(0.166) (0.173) (0.155) (0.165) (0.151) (0.107)
Constant 1.236** 1.588*** 0.225 1.483*** 0.611 3.770***
(0.511) (0.464) (0.432) (0.441) (0.449) (0.335)
Observations 782 807 834 860 960 2,382
Log Likelihood -512.243 -472.542 -541.840 -538.488 -605.132 -1,273.024
Akaike Inf. Crit. 1,042.486 963.084 1,101.679 1,094.975 1,228.264 2,564.047
Note: ***: p < 0.01; **: p < 0.05; *: p < 0.1

Appendix: Results of regression

Appendix: Predicted probabilities regarding satisfaction with democracy

Appendix: Predicted probabilities regarding diversity

References

1: https://online.hbs.edu/blog/post/tragedy-of-the-commons-impact-on-sustainability-issues

2: https://www.un.org/en/climatechange/cop27

3: https://www.un.org/en/climatechange/cop27

4: https://www.pewresearch.org/fact-tank/2021/05/26/key-findings-how-americans-attitudes-about-climate-change-differ-by-generation-party-and-other-factors/

5: https://www.pewresearch.org/science/2021/05/26/gen-z-millennials-stand-out-for-climate-change-activism-social-media-engagement-with-issue/

6: https://www.pewresearch.org/fact-tank/2021/05/26/key-findings-how-americans-attitudes-about-climate-change-differ-by-generation-party-and-other-factors/

7: https://climatetrade.com/which-countries-are-the-worlds-biggest-carbon-polluters/

8: https://ourworldindata.org/grapher/annual-change-fossil-fuels?tab=table

References: literature review

Age

Income

Political affiliation

References: literature review

Political affiliation

Education

Gender